| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9930394431554523, | |
| "eval_steps": 50, | |
| "global_step": 645, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02320185614849188, | |
| "grad_norm": 29.851993456423653, | |
| "learning_rate": 5e-07, | |
| "loss": 1.6448, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04640371229698376, | |
| "grad_norm": 14.665681394229273, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5147, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06960556844547564, | |
| "grad_norm": 7.818944779433003, | |
| "learning_rate": 9.998470286265414e-07, | |
| "loss": 1.2888, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09280742459396751, | |
| "grad_norm": 6.011350341653904, | |
| "learning_rate": 9.993882081071305e-07, | |
| "loss": 1.2291, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11600928074245939, | |
| "grad_norm": 5.834488035843988, | |
| "learning_rate": 9.986238191873872e-07, | |
| "loss": 1.1827, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13921113689095127, | |
| "grad_norm": 4.944851583820086, | |
| "learning_rate": 9.975543295858033e-07, | |
| "loss": 1.1637, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16241299303944315, | |
| "grad_norm": 5.051276519398371, | |
| "learning_rate": 9.961803937075514e-07, | |
| "loss": 1.1448, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.18561484918793503, | |
| "grad_norm": 5.008630441799495, | |
| "learning_rate": 9.945028522440653e-07, | |
| "loss": 1.1539, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2088167053364269, | |
| "grad_norm": 4.527889171881266, | |
| "learning_rate": 9.925227316586314e-07, | |
| "loss": 1.1302, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.23201856148491878, | |
| "grad_norm": 4.665448898349117, | |
| "learning_rate": 9.902412435583125e-07, | |
| "loss": 1.1315, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23201856148491878, | |
| "eval_loss": 1.1243596076965332, | |
| "eval_runtime": 107.3706, | |
| "eval_samples_per_second": 57.083, | |
| "eval_steps_per_second": 0.894, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2552204176334107, | |
| "grad_norm": 4.780072580854242, | |
| "learning_rate": 9.876597839525813e-07, | |
| "loss": 1.1184, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.27842227378190254, | |
| "grad_norm": 4.698899616021614, | |
| "learning_rate": 9.847799323991233e-07, | |
| "loss": 1.1112, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.30162412993039445, | |
| "grad_norm": 4.798544900134018, | |
| "learning_rate": 9.816034510373285e-07, | |
| "loss": 1.1022, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.3248259860788863, | |
| "grad_norm": 4.7554420584590105, | |
| "learning_rate": 9.781322835100637e-07, | |
| "loss": 1.1109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3480278422273782, | |
| "grad_norm": 4.715791460597178, | |
| "learning_rate": 9.743685537743856e-07, | |
| "loss": 1.1044, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.37122969837587005, | |
| "grad_norm": 4.781446446456838, | |
| "learning_rate": 9.70314564801922e-07, | |
| "loss": 1.0982, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.39443155452436196, | |
| "grad_norm": 4.616375479434296, | |
| "learning_rate": 9.659727971697173e-07, | |
| "loss": 1.0798, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.4176334106728538, | |
| "grad_norm": 4.81678899141071, | |
| "learning_rate": 9.613459075424033e-07, | |
| "loss": 1.0925, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4408352668213457, | |
| "grad_norm": 4.804913673447656, | |
| "learning_rate": 9.564367270466245e-07, | |
| "loss": 1.0726, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.46403712296983757, | |
| "grad_norm": 5.428814841421836, | |
| "learning_rate": 9.51248259538713e-07, | |
| "loss": 1.0732, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.46403712296983757, | |
| "eval_loss": 1.079745888710022, | |
| "eval_runtime": 106.1023, | |
| "eval_samples_per_second": 57.765, | |
| "eval_steps_per_second": 0.905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4872389791183295, | |
| "grad_norm": 4.842120094208203, | |
| "learning_rate": 9.457836797666721e-07, | |
| "loss": 1.0596, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.5104408352668214, | |
| "grad_norm": 5.09946277444771, | |
| "learning_rate": 9.400463314275941e-07, | |
| "loss": 1.0699, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5336426914153132, | |
| "grad_norm": 4.86963572334168, | |
| "learning_rate": 9.340397251217008e-07, | |
| "loss": 1.0653, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5568445475638051, | |
| "grad_norm": 4.762779682097982, | |
| "learning_rate": 9.27767536204258e-07, | |
| "loss": 1.0729, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.580046403712297, | |
| "grad_norm": 4.737687820279848, | |
| "learning_rate": 9.212336025366787e-07, | |
| "loss": 1.0709, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.6032482598607889, | |
| "grad_norm": 4.782611230029462, | |
| "learning_rate": 9.144419221381918e-07, | |
| "loss": 1.0579, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6264501160092807, | |
| "grad_norm": 4.991569479884324, | |
| "learning_rate": 9.073966507395121e-07, | |
| "loss": 1.0554, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.6496519721577726, | |
| "grad_norm": 5.008963657772678, | |
| "learning_rate": 9.001020992400085e-07, | |
| "loss": 1.043, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6728538283062645, | |
| "grad_norm": 5.0207483196868985, | |
| "learning_rate": 8.925627310699274e-07, | |
| "loss": 1.0531, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6960556844547564, | |
| "grad_norm": 5.288716673563479, | |
| "learning_rate": 8.84783159459285e-07, | |
| "loss": 1.0352, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6960556844547564, | |
| "eval_loss": 1.054144263267517, | |
| "eval_runtime": 106.0227, | |
| "eval_samples_per_second": 57.808, | |
| "eval_steps_per_second": 0.905, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7192575406032483, | |
| "grad_norm": 5.04033042010659, | |
| "learning_rate": 8.767681446150976e-07, | |
| "loss": 1.0565, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.7424593967517401, | |
| "grad_norm": 4.8573560396750395, | |
| "learning_rate": 8.68522590808682e-07, | |
| "loss": 1.0409, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.765661252900232, | |
| "grad_norm": 4.8296712724118125, | |
| "learning_rate": 8.600515433748001e-07, | |
| "loss": 1.0531, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7888631090487239, | |
| "grad_norm": 5.039044240861594, | |
| "learning_rate": 8.51360185624495e-07, | |
| "loss": 1.0321, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8120649651972158, | |
| "grad_norm": 4.944895975035593, | |
| "learning_rate": 8.424538356734956e-07, | |
| "loss": 1.0366, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8352668213457076, | |
| "grad_norm": 5.568497060519676, | |
| "learning_rate": 8.333379431881397e-07, | |
| "loss": 1.0454, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8584686774941995, | |
| "grad_norm": 4.984582777418547, | |
| "learning_rate": 8.240180860508026e-07, | |
| "loss": 1.0368, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.8816705336426914, | |
| "grad_norm": 5.7520355685162, | |
| "learning_rate": 8.144999669468713e-07, | |
| "loss": 1.0288, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9048723897911833, | |
| "grad_norm": 5.580835397673134, | |
| "learning_rate": 8.047894098753539e-07, | |
| "loss": 1.0323, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.9280742459396751, | |
| "grad_norm": 4.941555770840273, | |
| "learning_rate": 7.948923565852597e-07, | |
| "loss": 1.0347, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9280742459396751, | |
| "eval_loss": 1.0315721035003662, | |
| "eval_runtime": 106.1537, | |
| "eval_samples_per_second": 57.737, | |
| "eval_steps_per_second": 0.904, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.951276102088167, | |
| "grad_norm": 5.3026133930305335, | |
| "learning_rate": 7.848148629399285e-07, | |
| "loss": 1.0264, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.974477958236659, | |
| "grad_norm": 5.3528963387584945, | |
| "learning_rate": 7.745630952115363e-07, | |
| "loss": 1.0204, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9976798143851509, | |
| "grad_norm": 4.897176107855643, | |
| "learning_rate": 7.641433263080418e-07, | |
| "loss": 1.0287, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.0208816705336428, | |
| "grad_norm": 5.265360108783065, | |
| "learning_rate": 7.535619319348865e-07, | |
| "loss": 0.92, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0440835266821347, | |
| "grad_norm": 5.842663985054576, | |
| "learning_rate": 7.428253866937918e-07, | |
| "loss": 0.9119, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.0672853828306264, | |
| "grad_norm": 5.255941350708592, | |
| "learning_rate": 7.319402601210447e-07, | |
| "loss": 0.9091, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0904872389791183, | |
| "grad_norm": 5.099242988471265, | |
| "learning_rate": 7.209132126676933e-07, | |
| "loss": 0.8901, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.1136890951276102, | |
| "grad_norm": 5.963233610769237, | |
| "learning_rate": 7.097509916241145e-07, | |
| "loss": 0.8963, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.136890951276102, | |
| "grad_norm": 5.342236832387553, | |
| "learning_rate": 6.984604269914436e-07, | |
| "loss": 0.878, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.160092807424594, | |
| "grad_norm": 5.312138241916301, | |
| "learning_rate": 6.870484273023967e-07, | |
| "loss": 0.8776, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.160092807424594, | |
| "eval_loss": 1.0247548818588257, | |
| "eval_runtime": 105.9632, | |
| "eval_samples_per_second": 57.841, | |
| "eval_steps_per_second": 0.906, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1832946635730859, | |
| "grad_norm": 5.364600498172952, | |
| "learning_rate": 6.755219753940388e-07, | |
| "loss": 0.8925, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.2064965197215778, | |
| "grad_norm": 5.752065985447796, | |
| "learning_rate": 6.638881241350883e-07, | |
| "loss": 0.897, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2296983758700697, | |
| "grad_norm": 5.570817988636112, | |
| "learning_rate": 6.52153992110368e-07, | |
| "loss": 0.8839, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.2529002320185616, | |
| "grad_norm": 5.496660790853752, | |
| "learning_rate": 6.403267592650466e-07, | |
| "loss": 0.8916, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2761020881670533, | |
| "grad_norm": 5.463347675557559, | |
| "learning_rate": 6.28413662511334e-07, | |
| "loss": 0.8755, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2993039443155452, | |
| "grad_norm": 5.514605068924435, | |
| "learning_rate": 6.164219913003207e-07, | |
| "loss": 0.8775, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.322505800464037, | |
| "grad_norm": 5.727155908190926, | |
| "learning_rate": 6.043590831616676e-07, | |
| "loss": 0.9028, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.345707656612529, | |
| "grad_norm": 5.429710747722258, | |
| "learning_rate": 5.92232319213878e-07, | |
| "loss": 0.887, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.368909512761021, | |
| "grad_norm": 5.482889252366046, | |
| "learning_rate": 5.800491196478988e-07, | |
| "loss": 0.8759, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.3921113689095128, | |
| "grad_norm": 5.670067440966938, | |
| "learning_rate": 5.678169391868127e-07, | |
| "loss": 0.8888, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3921113689095128, | |
| "eval_loss": 1.0144710540771484, | |
| "eval_runtime": 105.8665, | |
| "eval_samples_per_second": 57.894, | |
| "eval_steps_per_second": 0.907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4153132250580047, | |
| "grad_norm": 5.530990774365388, | |
| "learning_rate": 5.555432625244023e-07, | |
| "loss": 0.8939, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.4385150812064964, | |
| "grad_norm": 5.572591226625499, | |
| "learning_rate": 5.432355997453728e-07, | |
| "loss": 0.8929, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4617169373549883, | |
| "grad_norm": 5.456192949381362, | |
| "learning_rate": 5.309014817300421e-07, | |
| "loss": 0.9025, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.4849187935034802, | |
| "grad_norm": 5.799863863147117, | |
| "learning_rate": 5.185484555463026e-07, | |
| "loss": 0.8753, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.5081206496519721, | |
| "grad_norm": 5.73261677068134, | |
| "learning_rate": 5.061840798316814e-07, | |
| "loss": 0.8938, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.531322505800464, | |
| "grad_norm": 5.762497249365714, | |
| "learning_rate": 4.938159201683186e-07, | |
| "loss": 0.882, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.554524361948956, | |
| "grad_norm": 5.533641700223646, | |
| "learning_rate": 4.814515444536974e-07, | |
| "loss": 0.8846, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.5777262180974478, | |
| "grad_norm": 5.782636301369744, | |
| "learning_rate": 4.69098518269958e-07, | |
| "loss": 0.8875, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.6009280742459397, | |
| "grad_norm": 5.461443407271733, | |
| "learning_rate": 4.5676440025462726e-07, | |
| "loss": 0.8713, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.6241299303944317, | |
| "grad_norm": 5.571957109149912, | |
| "learning_rate": 4.444567374755977e-07, | |
| "loss": 0.8902, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6241299303944317, | |
| "eval_loss": 1.0069366693496704, | |
| "eval_runtime": 106.2838, | |
| "eval_samples_per_second": 57.666, | |
| "eval_steps_per_second": 0.903, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6473317865429236, | |
| "grad_norm": 5.528554551947589, | |
| "learning_rate": 4.3218306081318713e-07, | |
| "loss": 0.8789, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.6705336426914155, | |
| "grad_norm": 5.640345642741927, | |
| "learning_rate": 4.199508803521012e-07, | |
| "loss": 0.8774, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6937354988399071, | |
| "grad_norm": 5.603129030745687, | |
| "learning_rate": 4.0776768078612207e-07, | |
| "loss": 0.8717, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.716937354988399, | |
| "grad_norm": 5.6707399409090895, | |
| "learning_rate": 3.9564091683833244e-07, | |
| "loss": 0.8764, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.740139211136891, | |
| "grad_norm": 5.448091702098461, | |
| "learning_rate": 3.835780086996793e-07, | |
| "loss": 0.865, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.7633410672853829, | |
| "grad_norm": 5.582199831694094, | |
| "learning_rate": 3.7158633748866607e-07, | |
| "loss": 0.8779, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.7865429234338746, | |
| "grad_norm": 5.6752676781492255, | |
| "learning_rate": 3.596732407349536e-07, | |
| "loss": 0.8672, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.8097447795823665, | |
| "grad_norm": 5.5587691039632015, | |
| "learning_rate": 3.4784600788963193e-07, | |
| "loss": 0.8725, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.8329466357308584, | |
| "grad_norm": 5.5052011161351615, | |
| "learning_rate": 3.3611187586491157e-07, | |
| "loss": 0.8615, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.8561484918793503, | |
| "grad_norm": 5.507497258082307, | |
| "learning_rate": 3.244780246059612e-07, | |
| "loss": 0.8773, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8561484918793503, | |
| "eval_loss": 0.9998582005500793, | |
| "eval_runtime": 105.9976, | |
| "eval_samples_per_second": 57.822, | |
| "eval_steps_per_second": 0.906, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8793503480278422, | |
| "grad_norm": 5.492700851735595, | |
| "learning_rate": 3.129515726976034e-07, | |
| "loss": 0.8671, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.902552204176334, | |
| "grad_norm": 5.7486034264931165, | |
| "learning_rate": 3.015395730085565e-07, | |
| "loss": 0.8614, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.925754060324826, | |
| "grad_norm": 5.124949765854869, | |
| "learning_rate": 2.902490083758856e-07, | |
| "loss": 0.871, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.948955916473318, | |
| "grad_norm": 5.408996022602446, | |
| "learning_rate": 2.790867873323067e-07, | |
| "loss": 0.8795, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.9721577726218098, | |
| "grad_norm": 5.750334709291967, | |
| "learning_rate": 2.680597398789554e-07, | |
| "loss": 0.8588, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.9953596287703017, | |
| "grad_norm": 5.711244881678974, | |
| "learning_rate": 2.5717461330620815e-07, | |
| "loss": 0.8719, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.0185614849187936, | |
| "grad_norm": 6.122014805843198, | |
| "learning_rate": 2.464380680651134e-07, | |
| "loss": 0.7933, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.0417633410672855, | |
| "grad_norm": 6.215562802835813, | |
| "learning_rate": 2.358566736919581e-07, | |
| "loss": 0.7943, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0649651972157774, | |
| "grad_norm": 5.90592154174332, | |
| "learning_rate": 2.2543690478846388e-07, | |
| "loss": 0.778, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.0881670533642693, | |
| "grad_norm": 5.689955483676048, | |
| "learning_rate": 2.1518513706007152e-07, | |
| "loss": 0.7847, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0881670533642693, | |
| "eval_loss": 1.0113998651504517, | |
| "eval_runtime": 106.2215, | |
| "eval_samples_per_second": 57.7, | |
| "eval_steps_per_second": 0.904, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.111368909512761, | |
| "grad_norm": 5.550398002087045, | |
| "learning_rate": 2.051076434147403e-07, | |
| "loss": 0.7843, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.1345707656612527, | |
| "grad_norm": 5.681734155354411, | |
| "learning_rate": 1.9521059012464607e-07, | |
| "loss": 0.7678, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.1577726218097446, | |
| "grad_norm": 5.865956254391275, | |
| "learning_rate": 1.855000330531289e-07, | |
| "loss": 0.7897, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.1809744779582365, | |
| "grad_norm": 6.055658613132172, | |
| "learning_rate": 1.7598191394919737e-07, | |
| "loss": 0.7811, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.2041763341067284, | |
| "grad_norm": 6.233987278809335, | |
| "learning_rate": 1.666620568118603e-07, | |
| "loss": 0.7975, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.2273781902552203, | |
| "grad_norm": 5.824376205357217, | |
| "learning_rate": 1.5754616432650443e-07, | |
| "loss": 0.7801, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.2505800464037122, | |
| "grad_norm": 6.375178064295241, | |
| "learning_rate": 1.4863981437550498e-07, | |
| "loss": 0.781, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.273781902552204, | |
| "grad_norm": 6.211607396754804, | |
| "learning_rate": 1.3994845662519983e-07, | |
| "loss": 0.7799, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.296983758700696, | |
| "grad_norm": 5.980327624770438, | |
| "learning_rate": 1.3147740919131812e-07, | |
| "loss": 0.783, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.320185614849188, | |
| "grad_norm": 5.925933936835478, | |
| "learning_rate": 1.2323185538490228e-07, | |
| "loss": 0.7668, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.320185614849188, | |
| "eval_loss": 1.0112136602401733, | |
| "eval_runtime": 105.6668, | |
| "eval_samples_per_second": 58.003, | |
| "eval_steps_per_second": 0.909, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.34338747099768, | |
| "grad_norm": 6.06031016994975, | |
| "learning_rate": 1.1521684054071523e-07, | |
| "loss": 0.7762, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.3665893271461718, | |
| "grad_norm": 6.173374303238529, | |
| "learning_rate": 1.0743726893007254e-07, | |
| "loss": 0.7806, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.3897911832946637, | |
| "grad_norm": 6.009209700505245, | |
| "learning_rate": 9.989790075999144e-08, | |
| "loss": 0.7731, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.4129930394431556, | |
| "grad_norm": 5.744648775110227, | |
| "learning_rate": 9.260334926048785e-08, | |
| "loss": 0.7892, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.4361948955916475, | |
| "grad_norm": 5.949296215452574, | |
| "learning_rate": 8.555807786180813e-08, | |
| "loss": 0.7785, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.4593967517401394, | |
| "grad_norm": 5.936014562230361, | |
| "learning_rate": 7.876639746332131e-08, | |
| "loss": 0.7829, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.4825986078886313, | |
| "grad_norm": 5.859713250927308, | |
| "learning_rate": 7.223246379574205e-08, | |
| "loss": 0.7919, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.505800464037123, | |
| "grad_norm": 6.243664538264793, | |
| "learning_rate": 6.596027487829913e-08, | |
| "loss": 0.7688, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.529002320185615, | |
| "grad_norm": 6.031972239373065, | |
| "learning_rate": 5.995366857240591e-08, | |
| "loss": 0.7789, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.5522041763341066, | |
| "grad_norm": 6.130593159142583, | |
| "learning_rate": 5.421632023332778e-08, | |
| "loss": 0.775, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.5522041763341066, | |
| "eval_loss": 1.0103154182434082, | |
| "eval_runtime": 105.6016, | |
| "eval_samples_per_second": 58.039, | |
| "eval_steps_per_second": 0.909, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.5754060324825985, | |
| "grad_norm": 5.963207804570064, | |
| "learning_rate": 4.8751740461286826e-08, | |
| "loss": 0.789, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.5986078886310904, | |
| "grad_norm": 5.8736607983865525, | |
| "learning_rate": 4.356327295337542e-08, | |
| "loss": 0.7992, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.6218097447795823, | |
| "grad_norm": 6.077763163177709, | |
| "learning_rate": 3.865409245759671e-08, | |
| "loss": 0.7824, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.645011600928074, | |
| "grad_norm": 5.999232007718555, | |
| "learning_rate": 3.402720283028277e-08, | |
| "loss": 0.7807, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.668213457076566, | |
| "grad_norm": 5.782915072578643, | |
| "learning_rate": 2.968543519807809e-08, | |
| "loss": 0.7836, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.691415313225058, | |
| "grad_norm": 5.895133763158038, | |
| "learning_rate": 2.5631446225614527e-08, | |
| "loss": 0.7866, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.71461716937355, | |
| "grad_norm": 5.789961425841594, | |
| "learning_rate": 2.1867716489936294e-08, | |
| "loss": 0.7724, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.737819025522042, | |
| "grad_norm": 5.666163619117144, | |
| "learning_rate": 1.8396548962671454e-08, | |
| "loss": 0.7903, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.7610208816705337, | |
| "grad_norm": 5.768500926901973, | |
| "learning_rate": 1.5220067600876684e-08, | |
| "loss": 0.7711, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.7842227378190256, | |
| "grad_norm": 6.087496150272388, | |
| "learning_rate": 1.2340216047418694e-08, | |
| "loss": 0.7768, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.7842227378190256, | |
| "eval_loss": 1.008864164352417, | |
| "eval_runtime": 105.8761, | |
| "eval_samples_per_second": 57.888, | |
| "eval_steps_per_second": 0.907, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8074245939675175, | |
| "grad_norm": 6.207391484980199, | |
| "learning_rate": 9.758756441687332e-09, | |
| "loss": 0.7875, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.8306264501160094, | |
| "grad_norm": 5.954613559365798, | |
| "learning_rate": 7.477268341368359e-09, | |
| "loss": 0.785, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.853828306264501, | |
| "grad_norm": 5.819327712838159, | |
| "learning_rate": 5.497147755934628e-09, | |
| "loss": 0.7832, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.877030162412993, | |
| "grad_norm": 5.786005516883902, | |
| "learning_rate": 3.819606292448541e-09, | |
| "loss": 0.7789, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.9002320185614847, | |
| "grad_norm": 6.130101975578697, | |
| "learning_rate": 2.4456704141967433e-09, | |
| "loss": 0.7973, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.9234338747099766, | |
| "grad_norm": 6.065019587560417, | |
| "learning_rate": 1.3761808126126483e-09, | |
| "loss": 0.7696, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.9466357308584685, | |
| "grad_norm": 5.601585670340757, | |
| "learning_rate": 6.117918928693622e-10, | |
| "loss": 0.7886, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.9698375870069604, | |
| "grad_norm": 6.062046858888585, | |
| "learning_rate": 1.529713734584326e-10, | |
| "loss": 0.7661, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.9930394431554523, | |
| "grad_norm": 6.1764579820644885, | |
| "learning_rate": 0.0, | |
| "loss": 0.7806, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.9930394431554523, | |
| "step": 645, | |
| "total_flos": 3802978389590016.0, | |
| "train_loss": 0.9239856169205304, | |
| "train_runtime": 9561.8566, | |
| "train_samples_per_second": 17.305, | |
| "train_steps_per_second": 0.067 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 645, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3802978389590016.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |